当前位置:首页 >> 经济学 >>

Financial development and economic growth


Review of Financial Economics 11 (2002) 131 – 150

Financial development and economic growth Another look at the evidence from developing countries
Yousif Khalifa Al-Yousif *
Department of Economics, United Arab Emirates University, P.O. Box 25306, Abu-Dhabi, United Arab Emirates Received 4 January 2001; received in revised form 5 July 2001; accepted 4 February 2002

Abstract The present paper examines the nature and direction of the relationship between financial development and economic growth using both time-series and panel data from 30 developing countries for the period 1970– 1999. The choice of the sample was determined by the availability of data. As such, the exclusion of other developing countries is due to the fact that the data on these countries are missing for some years. The empirical results strongly support the view that financial development and economic growth are mutually causal, that is, causality is bidirectional. There is also some support for the other views presented in the literature (supply-leading, demand-leading, and no relationship) but it is not as strong as that for the bidirectional causality. Moreover, the findings of the present paper accords with the view of the World Bank and other empirical studies that the relationship between financial development and economic growth cannot be generalized across countries because economic policies are country specific and their success depends, among others things, on the efficiency of the institutions implementing them. D 2002 Elsevier Science Inc. All rights reserved.
JEL classification: O16; G18; G28 Keywords: Financial development; Economic growth

* Fax: +971-2-4436376. E-mail address: subhanah@emirates.net.ae (Y. Khalifa Al-Yousif). 1058-3300/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 8 - 3 3 0 0 ( 0 2 ) 0 0 0 3 9 - 3

132

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

1. Introduction Ever since the pioneering works of Goldsmith (1969) and Shumpeter (1932) and more recently of McKinnon (1973) and Shaw (1973), the relationship between financial development1 and economic growth has been one important area of discussion among economists. Later studies, both theoretical and empirical, have attempted to deepen our understanding of the different aspects of this relationship by exploring the existence of this relationship, the direction of causality between the two variables, and the channels of transmission between them. A number of recent papers have surveyed this literature (see, for example, DemirgucKunt & Levine, 1996; Levine, 1997; Thakor, 1996). Therefore, rather than presenting another review, in the present paper, I will briefly discuss the main streams of thought emanating from this literature on the relationship between financial development and economic growth by way of introduction to the empirical analysis to follow. The above mentioned literature indicates that economists hold different views on the existence and direction of causality between financial development and economic growth. The first is the ‘‘supply-leading’’ view, which states that financial development has a positive effect on economic growth. According to this view, financial intermediation contributes to economic growth through two main channels: (1) by raising the efficiency of capital accumulation and in turn the marginal productivity of capital (Goldsmith, 1969) and (2) by raising the savings rate and thus the investment rate (McKinnon, 1973; Shaw, 1973). In other words, by increasing the size of savings and improving the efficiency of investment, financial development leads to higher economic growth. This first view has received considerable support from recent empirical studies (see, for example, Bencivenga & Smith, 1991; Greenwood & Jovanovic, 1990; Thakor, 1996). The second view of the relationship between the two variables was advanced by Robinson (1952) and it states that financial development follows economic growth or ‘‘where enterprise leads finance follows’’ (Robinson, 1952, p. 86). According to this ‘‘demand-following’’ view, as the real side of the economy expands, its demand for financial services increases, leading to the growth of these services. Empirical support for this second view can also be found in some recent studies (Demetrides & Hussein, 1996; Friedman & Schwartz, 1963; Ireland, 1994). A third view of the relationship between financial development and economic growth postulates that the two variables are mutually causal, that is, they have a bidirectional causality (Demetrides & Hussein, 1996; Greenwood & Smith, 1997). Finally, a fourth view, which was originally put forward by Lucas (1988), argues that financial development and economic growth are not causally related or in the words of Lucas, ‘‘economists badly overstress the role of financial factors in economic growth.’’ It is obvious from the preceding brief exposition of the different streams of thought on the relationship between financial development and economic growth that the literature is mixed
The terms ‘‘financial development’’ and ‘‘financial intermediation’’ are used interchangeably in the present paper, although the former is broader and includes capital markets in addition to financial intermediaries. However, because of the lack of data on capital markets in the countries studied in this paper, the two terms are used interchangeably.
1

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

133

and inconclusive. This can be attributed to a number of reasons. One is that most of the current literature has focused on the correlation between financial development and economic growth. However, correlation does not always mean causality. Therefore, putting more emphasis on the causality issue is more helpful in understanding the issues at hand. Another weakness in many of the previous studies is the use of an augmented production function approach where a measure of financial development is added as another argument in the production function (Wang, 1999). The problem with this approach is that it assumes that economic growth is the dependent variable. Thus, causality runs from financial development to economic growth. However, as mentioned earlier, there are also strong theoretical arguments for the possibility that this causality runs from economic growth to financial development (the demand-following hypothesis). As such, these studies suffer from the problem of model misspecification. Moreover, much of the existing literature uses crosssection data, which do not resolve the issue of causality. Therefore, the present paper examines the finance-growth nexus in 30 developing countries2 using both time-series and panel data and employing Granger-causality tests within an error-correction (EC) framework. The rest of the paper is organized as follows. Section 3.1 reports the empirical results from unit root tests and discusses the results from the Johansen test of cointegration. Sections 3.2 and 3.3 analyze the Granger-causality tests for both time-series and panel data and their implications. Section 4 concludes.

2. Methodology and data description To explore the relationship between financial development and economic growth, I use a Granger-causality test within an EC model (ECM) (Bishop, 1979). This test, however, requires that the variables used in a given model be stationary, that is, their stochastic properties are time invariant. Many studies have shown that models with nonstationary variables tend to produce spurious regressions and make the usual test statistics (t, F, DW, and R2) unreliable (see, for example, Granger & Newbold, 1974; Stock & Watson, 1989). If differenced appropriately, however, a nonstationary variable can achieve stationarity (Granger, 1986). The appropriate number of differencing is called the order of integration. Hence, if a time-series Z becomes stationary after being differenced d times, we say that Z is integrated of order d, denoted by Z $ I(d). To find the proper order of integration for the two variables in the present model, I use two testing procedures: Augmented Dickey Fuller (ADF) test and the Perron–Phillips (PP) test (see, for example, Enders, 1995; Pantula, Gonzalo-Farias, & Fuller, 1994). However, differencing the variables of the model to achieve stationarity can dissipate long-run information if the these variables have a long-run relationship, that is, if they are cointegrated. Thus, if a model with stationary variables is estimated without regard to the
These countries include Saudi Arabia, Kuwait, Egypt, Oman, Bahrain, Iran, Qatar, Tunisia, Morocco, UAE, Philippines, Thailand, Ghana, Venezuela, Cyprus, Korea, Sri Lanka, South Africa, Niger, Guatemala, Madagascar, Haiti, Pakistan, Singapore, Syria, Tanzania, Kenya, Trinidad and Tobago, Jordan, and Malaysia.
2

134

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 1 Unit root test results: the Johansen – Juselius approach Country Saudi Arabia Variables GP C F DGB DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGB DC DF GP C F DGP DC DF GP C F DGP ADF (L) ? 0.72 0.05 ? 2.17 ? 1.67 ? 2.19 ? 2.68 ? 0.78 ? 0.13 ? 2.39 ? 2.92 ? 3.45 ? 3.17 ? 2.67 ? 1.61 ? 3.20 ? 2.40 ? 2.44 ? 3.25 1.62 0.52 ? 1.96 ? 2.98 ? 3.17 ? 3.25 0.57 0.39 ? 3.05 ? 1.07 ? 3.04 ? 2.77 ? 0.31 1.83 0.53 ? 2.03 ? 3.53 ? 1.81 ? 4.61 ? 0.98 ? 1.63 ? 1.03 ? 2.49 ? 2.05 2.15 ? 2.50 ? 2.21 ? 1.01 (3) (3) (4) (3)* (2)* (3)** (2) (2) (2) (3)* (2)* (2)* (2) (2) (2) (2)* (2)* (3)** (4) (4) (4) (2)* (4)** (3)** (2) (2) (2) (4) (2)** (4)* (2) (4) (3) (4)* (3)* (3)* (2)* (2) (2) (4) (2)** (2)** (2) (3) (4) (2) PP (L) ? 0.06 (3) 0.30 (3) ? 1.94 (4) ? 11.97 (3)** ? 12.14 (2)** ? 11.59 (3)** ? 0.09 (2) ? 0.02 (2) ? 4.27 (2) ? 21.10 (3)** ? 26.73 (2)** ? 20.64 (2)** ? 7.45 (2) ? 5.95 (2) ? 3.34 (2) ? 17.22 (2)** ? 28.68 (2)** ? 12.87 (3)** 0.08 (4) ? 0.06 (4) ? 0.64 (4) ? 30.62 (2)** ? 12.14 (4)** ? 12.87 (3)** ? 0.01 (2) ? 0.05 (2) ? 1.85 (2) ? 13.80 (4)** ? 19.42 (2)** ? 28.20 (4)** ? 0.01 (3) 0.37 (4) ? 0.93 (4) ? 15.22 (4)** ? 25.32 (3)** ? 13.46 (2)** ? 1.80 (2) ? 0.15 (2) ? 2.60 (2) ? 16.15 (4)** ? 26.19 (2)** ? 14.33 (2)** 0.07 (2) ? 5.70 (3) ? 6.93 (4) ? 9.93 (2)* (continued on next page)

Kuwait

Egypt

Oman

Bahrain

Iran

Qatar

Tunisia

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 1 (continued ) Country Tunisia Morocco Variables DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C ADF (L) ? 2.91 (2)* ? 1.64 (2)* 1.40 (3) 1.38 (2) ? 1.11 (4) ? 2.60 (2)** ? 1.34 (2)** ? 0.72 (3)** ? 0.95 (2) ? 0.49 (2) ? 2.63 (2) ? 2.06 (2)* ? 2.93 (2)* ? 1.79 (3)* ? 1.75 (2) ? 0.55 (2) 1.39 (2) ? 2.78 (2)** ? 3.97 (2)* ? 3.22 (2)* 0.12 (4) ? 0.85 (2) ? 1.35 (2) ? 3.12 (3)* ? 1.69 (2)* ? 1.91 (2)* ? 0.88 (2) 1.31 (3) 1.59 (3) ? 3.50 (2) ? 1.89 (2)* ? 3.29 (2)* 2.66 (4) 0.42 (4) ? 1.04 (3) ? 1.74 (2)* ? 2.54 (3)* ? 0.18 (2) 1.66 (2) ? 0.47 (2) ? 1.01 (2) ? 2.88 (3)** ? 3.46 (2)** ? 3.80 (4)** 1.19 (2) 2.48 (2) PP (L) ? 15.27 ? 17.86 0.03 0.24 ? 0.89 ? 37.76 ? 29.37 ? 35.08 ? 0.02 ? 0.15 ? 1.21 ? 20.96 ? 22.76 ? 19.41 ? 0.08 ? 0.29 ? 0.09 ? 30.20 ? 21.92 ? 29.55 0.045 ? 0.39 ? 0.81 ? 17.11 ? 23.51 ? 17.59 0.49 5.30 ? 1.33 ? 20.42 ? 29.33 ? 17.43 0019 ? 0.08 ? 0.70 ? 5.61 ? 27.65 ? 12.14 0.09 0.06 ? 0.49 10.80 27.53 19.42 0.01 0.52

135

UAE

Venezuela

Philippines

Cyprus

Korea

Sri Lanka

South Africa

(2)* (2)* (3) (2) (4) (2)** (2)** (3)** (2) (2) (2) (2)** (2)** (3)** (2) (2) (2) (2)** (2)** (2)** (4) (2) (2) (3)** (2)** (2)** (2) (3) (3) (2)** (2)** (2)* (4) (4) (3) (2) (3)** (2)* (2) (2) (2) (3)* (2)** (4)** (2) (2)

(continued on next page)

136

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 1 (continued ) Country South Africa Variables F DGB DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF ADF (L) ? 0.00 2.55 ? 1.96 ? 2.28 ? 1.09 ? 0.56 0.17 ? 2.66 ? 2.86 ? 2.08 ? 0.06 ? 0.18 0.20 ? 2.35 ? 3.58 ? 3.40 ? 0.49 0.51 ? 0.07 ? 3.54 ? 3.62 ? 2.64 ? 1.04 ? 0.77 ? 0.83 ? 2.77 ? 4.00 ? 2.35 ? 2.55 ? 2.24 ? 1.63 ? 2.17 ? 2.77 ? 1.97 3.48 0.37 ? 0.88 ? 0.73 ? 2.83 ? 3.31 ? 0.05 0.01 ? 1.18 ? 2.30 ? 2.79 ? 2.40 (3) (4) (2)* (2)* (2) (4) (2) (2) (3)** (2)* (3) (2) (4) (2) (4)** (4)* (2) (2) (2) (3)** (3)** (2)** (2) (3) (3) (2)* (2)** (2)* (4) (3) (2) (3) (2)* (2) (2) (4) (4) (4) (4)** (3)** (3) (2) (4) (2)* (2)* (4)* PP (L) 0.02 (3) ? 17.47 (4)** ? 18.97 (2)** ? 21.47 (2)** ? 0.07 (2) ? 0.05 (4) 0.12 (2) ? 21.24 (2)** ? 26.56 (3)** ? 19.44 (2)** 0.02 (3) 0.09 (2) 0.08 (4) ? 15.46 (2) 21.9 (4) ? 24.56 (4)** ? 0.02 (2) 0.12 (2) ? 0.00 (2) ? 17.57 (3)** ? 27.49 (2)** ? 22.03 (2)** ? 0.03 (2) ? 1.91 (3) ? 0.57 (3) ? 24.77 (2)** ? 32.39 (2)** ? 30.29 (2)** ? 6.03 (4) ? 7.33 (3) ? 3.67 (2) ? 17.23 (3)* ? 20.47 (2)* ? 21.91 (2)** 0.08 (2) 0.17 (4) ? 0.25 (4) ? 16.90 (4)** ? 23.54 (4)** ? 19.66 (3)** ? 0.00 (3) 0.02 (2) ? 1.00 (4) ? 15.38 (2)** ? 15.38 (2)** ? 14.25 (4)** (continued on next page)

Niger

Guatemala

Madagascar

Haiti

Ghana

Pakistan

Syria

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 1 (continued ) Country Tanzania Variables GP C F DGP DC DF GP C F DGP DC DF GP C F DGB DC DF GP C F DGP DC DF GP C F DGP DC DF GP C F DGP DC DF DDGP DDC DDF GP C F DGP DC DF ADF (L) ? 0.30 (2) ? 2.39 (2) ? 1.09 (4) ? 4.52 (2)** ? 2.60 (2)** ? 0.29 (2) ? 0.57 (2) ? 0.93 (2) ? 0.67 (2) ? 2.11 (2)* ? 3.63 (2)** ? 2.41 (2)* ? 0.89 (3) 0.61 (2) ? 1.42 (2) ? 1.65 (2)* ? 3.09 (2)** ? 2.30 (3)* ? 0.16 (3) ? 0.47 (2) ? 1.57 (4) ? 2.66 (2)** ? 3.31 (2)** ? 1.77 (3)* 2.15 (2) ? 2.50 (3) ? 2.21 (4) ? 1.01 (2) ? 2.91 (2)* ? 1.64 (2)* 1.55 (4) 0.26 (4) ? 2.55 (2) 1.55 (4) 0.26 (4) ? 2.55 (2) ? 1.71 (4)* ? 1.83 (4)* ? 1.91 (2)* ? 2.08 (3) ? 1.94 (2) ? 3.31 (4) ? 3.44 (4)* ? 2.62 (2) ? 2.46 (3) PP (L) ? 0.40 ? 0.43 ? 3.10 ? 33.16 ? 34.11 ? 25.12 ? 0.04 ? 0.36 ? 0.39 ? 19.66 ? 27.46 ? 29.16 ? 0.03 0.25 ? 0.96 ? 12.09 ? 21.66 ? 22.42 0.04 0.01 ? 3.80 ? 16.87 ? 23.20 ? 21.74 0.07 ? 5.70 ? 6.93 ? 9.93 ? 15.27 ? 17.86 0.10 0.30 ? 1.93 0.10 0.30 ? 1.93 ? 15.13 ? 14.58 ? 20.90 ? 4.21 ? 4.65 ? 9.28 ? 15.99 ? 26.07 ? 25.10

137

Kenya

Trinidad and Tobago

Jordan

Tunisia

Malaysia

Singapore

(2) (3) (2) (2)** (2)** (2)** (2) (2) (2) (2)** (2)** (2)** (3) (2) (2) (2)* (2)** (2)** (3) (2) (4) (2)** (2)** (3)** (2) (3) (4) (2)* (2)* (2)* (4) (4) (2) (4) (4) (2) (4)** (4)** (2)** (3) (2) (4) (4)* (2)* (2)*

(continued on next page)

138

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 1 (continued ) Country Thailand Variables GP C F DGP DC DF DDGP DDC DDF ADF (L) 1.16 ? 0.47 ? 1.84 ? 1.20 ? 2.30 ? 1.28 ? 3.73 ? 2.84 ? 4.29 (4) (2) (3) (2) (2) (2) (2)** (2)** (2)** PP (L) 0.23 (4) ? 0.22 (2) ? 1.57 (3) ? 3.57 (2) ? 27.44 (2) ? 8.11 (2) ? 20.96 (2)** ? 37.55 (2)** ? 24.94 (2)**

All variables are in natural logs. L is the optimal lag according to the Akaiki Information Criteria (AIC). DD is the second-difference operator. * Indicates rejection of the null hypothesis at the 10% level of significance. ** Indicates rejection of the null hypothesis at the 5% level significance.

cointegration between them, then this model is also inappropriate because it suffers from an omitted variable bias. Again, to deal with this problem, Engel and Granger (1987) show that a system of cointegrated variables can be represented by a dynamic ECM. This model is formed by adding an EC term as another regressor to the model with the stationary variables. This term is the lagged-once residuals that are generated from the cointegrating relationship between the variables in the model. To some researchers, the EC term represents the longrun Granger-causality. Therefore, before examining the causal relationship between financial development and economic growth, I begin by converting the two variables to stationary time-series and then test if they are cointegrated. To that end, I employ the Johansen (1988) efficient maximumlikelihood approach. Finally, a word about the measurement of the variables and the data used is in order. Following the common practice in the literature, I measure economic growth by the growth rate of per capita real GDP denoted in the rest of the paper by GP. Two proxies are used for financial development. One is the currency ratio, denoted by C, and calculated as the ratio of currency/narrow money stock (M1). A fall in this ratio indicates a higher diversification of financial institutions and a greater availability and use of noncurrency transaction methods. According to Vogel and Buser (1976), this proxy for financial development can assess the complexity of domestic financial markets. The second proxy of financial development is the inverse of the broad-money velocity, that is, the ratio of broad money stock (M2)/nominal GDP. This measure was put forward by McKinnon (1973) and Shaw (1973) and recently used by King and Levine (1993). This measure, which is often called the monetization variable, is denoted by F in this paper and used as a proxy for the size of the market. An increase in F implies an expansion in the financial intermediary sector relative to the rest of the economy. The empirical results in this paper are derived from annual time-series and panel data on real GDP per capita and the two proxies for financial development (C and F) over the period 1970–1999. The data on per capita real GDP, currency, narrow money, and broad money are

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

139

from different issues of the IMF International Financial Statistics (IFS). Finally, it is worth noting that a number of developing countries were excluded from our sample because of the lack of data on one or more of the variables included in the model.

3. Empirical results 3.1. Unit root and cointegration test results Table 1 displays the results from the unit root test. The evidence in the table indicates that according to both the ADF and the PP procedures, the three variables (GP, C, and F) are stationary at first differences in 28 out of the 30 countries in our sample and it is stationary at second differences in the remaining two. Having established the stationarity of the three variables, the next step is to test for possible cointegration between them. The results from the Johansen test are shown in Table 2. These results show that real GDP (GP) per capita is cointegrated with the currency ratio (C) in 17 out of the 30 countries and it is cointegrated with the monetization measure ( F) in 28 countries. Accordingly, an EC term will be added to the causality test equations to capture this longrun relationship when it exists between a set of variables. 3.2. Granger-causality test results from time-series data To test whether financial development Granger-causes economic growth, I estimate the following bivariate ECM for each country:
n1 X i?1 n2 X i?1

DGP ? a0 ?

a1i DGPt?i ?

a2i DMt?i ? yECt?1 ? et

?1?

where the two variables are expressed in first differences (D) as determined by the stationarity test, GP denotes real gross domestic product, M denotes the measure of financial development, which is either the currency ratio C or the monetization variable F, EC is the EC term taken from the bivariate cointegrating relationship, e is the white-noise error term, t denotes time in years, and the n’s are the lag orders of the a’s. I use the Hendry General-to-Specific modeling strategy to determine the proper lag length on each independent variable (Gilber, 1986). The null hypothesis that financial development does not Granger-cause economic growth is rejected if the coefficients on the distributed-lagged financial development variables (a2i’s) are found to be statistically significant as a group and/or the coefficient on the EC term is found to be statistically significant. The significance of a2i indicates short-run Granger-causality, while the significance of the j coefficient indicates long-run Granger-causality between the two variables. Of course, if the two variables were not cointegrated, the EC term will not appear in the above equation, in which case we test only for short-run causality.

140

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 2 Cointegration test results The Johansen (trace) test Country Saudi Arabia Cointegrating vector (GP,C) Cointegrating vector (GP,F) Kuwait Cointegrating vector (GP,C) Cointegrating vector (GP,F) Egypt Cointegrating vector (GP,C) Cointegrating vector (GP,F) Oman Cointegrating vector (GP,C) Cointegrating vector (GP,F) Bahrain Cointegrating vector (GP,C) Cointegrating vector (GP,F) Iran Cointegrating vector (GP,C) Cointegrating vector (GP,F) Qatar Cointegrating vector (GP,C) Cointegrating vector (GP,F) Tunisia Cointegrating vector (GP,C) Cointegrating vector (GP,F) Morocco Cointegrating vector (GP,C) Null hypothesis r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 Test statistics 27.52** 2.58 29.54** 5.40 13.73 3.06 12.81 5.77 12.06 3.05 26.06** 5.07 28.87** 12.42 17.80* 7.09 15.07 2.79 8.11 3.66 19.00* 4.13 15.44 5.39 15.10 5.17 30.00** 5.83 53.89** 2.39 72.48** 6.60 49.51** 4.18 (continued on next page)

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 2 (continued ) The Johansen (trace) test Country Morocco Cointegrating vector (GP,F) UAE Cointegrating vector (GP,C) Cointegrating vector (GP,F) Venezuela Cointegrating vector (GP,C) Cointegrating vector (GP,F) Philippines Cointegrating vector (GP,C) Cointegrating vector (GP,F) Cyprus Cointegrating vector (GP,C) Cointegrating vector (GP,F) Korea Cointegrating vector (GP,C) Cointegrating vector (GP,F) Sri Lanka Cointegrating vector (GP,C) Cointegrating vector (GP,F) South Africa Cointegrating vector (GP,C) Cointegrating vector (GP,F) Niger Cointegrating vector (GP,C) Cointegrating vector (GP,F) Null hypothesis r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1

141

Test statistics 37.64** 10.28 15.00 3.98 18.21* 1.19 11.19 1.98 14.82 5.88 23.52* 4.28 18.31* 6.10 9.41 4.05 31.84** 4.66 20.81* 5.11 46.32** 6.36 20.72* 6.29 32.77** 8.83 24.74* 3.70 16.67 5.69 8.35 1.27 5.59 1.91 (continued on next page)

142

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 2 (continued ) The Johansen (trace) test Country Guatemala Cointegrating vector (GP,C) Cointegrating vector (GP,F) Madagascar Cointegrating vector (GP,C) Cointegrating vector (GP,F) Haiti Cointegrating vector (GP,C) Cointegrating vector (GP,F) Ghana Cointegrating vector (GP,C) Cointegrating vector (GP,F) Pakistan Cointegrating vector (GP,C) Cointegrating vector (GP,F) Syria Cointegrating vector (GP,C) Cointegrating vector (GP,F) Tanzania Cointegrating vector (GP,C) Cointegrating vector (GP,F) Kenya Cointegrating vector (GP,C) Cointegrating vector (GP,F) Trinidad and Tobago Cointegrating vector (GP,C) Cointegrating vector (GP,F) Null hypothesis r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 Test statistics 13.25 3.99 20.21* 3.17 15.75 6.69 19.56* 4.29 15.84 1.17 21.48* 6.76 19.68* 5.25 18.96* 3.52 25.68* 6.32 36.18** 15.45 18.08* 3.92 20.13** 3.55 25.48** 1.70 22.04** 1.15 9.06 3.83 5.85 2.79 10.58 2.56 19.26* 5.43 (continued on next page)

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 2 (continued ) The Johansen (trace) test Country Jordan Cointegrating vector (GP,C) Cointegrating vector (GP,F) Malaysia Cointegrating vector (GP,C) Cointegrating vector (GP,F) Singapore Cointegrating vector (GP,C) Cointegrating vector (GP,F) Thailand Cointegrating vector (GP,C) Cointegrating vector (GP,F) Null hypothesis r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1 r=0 r 1

143

Test statistics 18.17* 4.25 27.92** 5.04 21.02** 7.43 27.51** 7.05 44.61* 17.23 22.02** 7.67 59.28** 3.10 17.72 7.55

r denotes the number of cointegrating vectors. * Indicates rejection of the null hypothesis at the 10% level of significance. ** Indicates rejection of the null hypothesis at the 5% level significance.

To test the opposite hypothesis that economic growth Granger-causes financial development, I estimate another equation similar to the first equation, except that now financial development will be the dependent variable. Therefore, the second equation will be as follows: DMt ? b0 ?
m1 X i?1

b1i DMt?i ?

m2 X i?1

b2i DGPt?i ? fECt?1 ? mt :

?2?

As in Eq. (1), our focus is on the statistical significance of the group coefficients b2i’s or short-run causality and on the statistical significance of f, which represents long-run causality. The empirical results from Eqs. (1) and (2) for the 30 countries are reported in Table 3. These results paint a mixed picture about the relationship between financial development and economic growth. For example, when the currency ratio C is employed as a measure of financial development, the results indicate that in 10 out of the 30 countries the causality runs from per capita real GDP to financial development, in 8 countries the causality is bidirectional, in 4 countries financial development Granger-causes per capita real GDP, and in the remaining 8 countries financial development and economic growth are not related. Also, when causality exists, it may be a short-run, a long-run, or both a short-run and a long-run phenomenon.

144

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 3 Granger-causality test results from ECMs Null hypothesis Saudi Arabia C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Kuwait C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Egypt C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Oman C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Bahrain C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Iran C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Qatar C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Tunisia C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Morocco C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Short-run 0.39 2.32 1.89* 9.96** 0.39 0.36 0.76 0.64 1.45 1.88* 0.51 0.56 1.41 0.18 0.67 0.01 2.32** 0.45 1.38 2.13** 0.69 2.47** 0.79 2.49** 0.60 0.29 0.70 0.55 1.10 0.36 1.43 0.36 0.12 0.97 1.27 1.64* Long-run 1.64* 3.03* 2.98** 1.91* 0.31 0.42 0.39 0.23 0.89 1.57 3.17** 3.01** 3.87** 3.46** 4.24 1.76* – – – – 1.54* 0.45 2.75** 0.38 – – 0.58 2.29** 0.15 2.06** 0.35 2.50** 0.79 0.75 1.13 0.26 (continued on next page)

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 3 (continued ) Null hypothesis UAE C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Venezuela C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Philippines C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Cyprus C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Korea C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Sri Lanka C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F South Africa C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Niger C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Guatemala C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Short-run 0.04 0.87 2.09 0.24 0.51 0.28 0.98 0.24 0.15 1.21 1.67* 0.67 2.97** 1.09 2.17** 0.21 0.16 0.94 1.52 1.08 1.39 0.72 0.53 0.67 0.01 1.19 1.00 0.27 0.34 0.77 1.32 1.02 0.73 0.41 0.77 0.93

145

Long-run – – 2.23** 0.82 – – – – 2.92** 1.55* 3.07** 0.39 – – 0.69 2.03** 2.82** 1.88* 2.25** 0.34 1.92* 2.15** 0.20 3.69** 2.74** 0.77 – – – – – – – – 1.31 2.95** (continued on next page)

146

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Table 3 (continued ) Null hypothesis Madagascar C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Haiti C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Ghana C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Pakistan C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Syria C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Tanzania C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Kenya C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Trinidad and Tobago C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Jordan C does not Granger-cause GP GP does not Granger-cause C F does not Granger-cause GP GP does not Granger-cause F Short-run 2.96** 2.41** 0.04 0.74 0.96 1.12 0.03 0.61 0.29 0.11 0.88 1.08 1.35 0.61 0.13 1.45 0.37 1.22 1.59 1.06 0.55 1.83* 0.41 1.65* 1.45 1.05 0.78 0.19 0.63 0.61 1.10 0.01 1.32 0.49 0.66 0.96 Long-run – – 2.18** 1.99** 1.28 3.91** – – 2.54** 2.25** 1.97** 2.12** 0.92 2.29** 0.34 4.37** 0.01 4.22** 2.62** 0.90 4.23 0.01 3.98** 1.58 – – – – 1.77* 1.76* 2.45** 0.41 1.63* 1.39 2.05* 2.21** (continued on next page)

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150 Table 3 (continued ) Null hypothesis Short-run

147

Long-run

Malaysia C does not Granger-cause GP 1.52 0.91 GP does not Granger-cause C 0.45 1.93* F does not Granger-cause GP 1.36 2.65** GP does not Granger-cause F 2.89** 4.71** Singapore C does not Granger-cause GP 0.57 1.09 GP does not Granger-cause C 1.73* 1.43 F does not Granger-cause GP 0.58 2.16** GP does not Granger-cause F 2.08** 1.83* Thailand C does not Granger-cause GP 0.39 0.39 GP does not Granger-cause C 0.96 3.27** F does not Granger-cause GP 0.11 – GP does not Granger-cause F 1.58 – The statistics in the short-run denote t statistics if the independent variable is lagged once and denote F-statistics if it is lagged more than once. The long-run statistics are the t statistics for the EC term if the variables are cointegrated, while a straight line ( – ) indicates no cointegration. * Indicates rejection at the 10% level of significance. ** Indicates rejection of the null hypothesis at the 5% level of significance.

Similarly, when we use the monetization variable as a proxy for financial development, the results are also mixed. In 10 of the 30 countries in our sample, causality between financial development and economic growth is bidirectional, in 9 countries it runs from economic growth to financial development, in 4 countries it runs from financial development to economic growth, and in the remaining 7 countries there does not seem to be a significant relationship between the two variables in question. Finally, although most of the results in this paper point to a positive correlation between the two variables, there are some cases where a negative correlation is found between these two variables. This finding, which seems puzzling at first glance, has been reported in other studies (e.g. Gertler & Rose, 1991; Gregorio & Guidotti, 1995). One possible explanation for this negative correlation is that it is a result of the business cycle rather than a representation of a long-run relationship. Another explanation is that it is due to the fact that financial intermediaries are operating in a weak regulatory environment combined with the expectation that governments will bail out failing banks. As a result, these financial institutions were inefficient in their allocation of resources. This inefficiency may lead to a reduction in the rate of economic growth. In my view, both of these forces were at work in these countries because many of the countries in our sample experienced long business cycles in the 1980s and mid1990s. Also, they suffer from fragile financial systems and weak regulatory environment, which has led to a number of financial crises during the 1980s and early 1990s. Examples include the banking crises in Egypt, the crash of the Al-Manakh capital market in Kuwait, and the bankruptcy of the BCCI, which was owned by the UAE and the East Asian crisis of 1997–1998.

148

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

3.3. Granger-causality test results from panel data In addition to the above results, which where derived from time-series data for 30 countries over the period 1977–1999, I use panel techniques in which cross-section and time-series data are pooled together to test the nature of the relationship between financial development and economic growth. The empirical results from using this technique show that the three variables (GP, C, and F) are stationary at first differences and that the two measures of financial development (C and F) have a long-run relationship (cointegrated) with real GDP per capita (GP). As with regard to the causality issue, these results indicate that there is a bidirectional causality between real GDP per capita and each of the two proxies of financial development. This is supportive of the results obtained from time-series data especially when financial development is measured by the broad monetization variable.3

4. Conclusions and policy implications This paper examines the relationship between financial development and economic growth in 30 developing countries using a Granger-causality test within an EC framework. The empirical findings and their policy implications can be summarized as follows. First, there is strong support from both time-series and panel data for the view that the causality between financial development and economic growth is a bidirectional one. Second, there is also some support for other views including the ‘‘supply-leading,’’ the ‘‘demand-leading,’’ and the view that there is no relationship between the two variables although this support is not as strong as the one for the ‘‘bidirectional’’ view. Third, our findings also show that the results are country specific and tend to vary with the kind of proxies used to measure financial development. This can be attributed to the fact that these countries differ in their level of financial development due to differences in policies and institutions. These results accord with the view of the World Bank that economic policies are country specific and their success is a function of the institutions that implement them (World Bank, 1993). Other studies (see, for example, Darrat, 1999; Demetrides and Hussein, 1996) have also reported results that are country specific with a high degree of variation across measures of financial development. Finally, some of the results of the present paper show a negative correlation between financial development and economic growth. These results were also reported in other studies as I mentioned earlier. In addition, they can be attributed to the business cycle that these countries have experienced during the 1980s and/or to the weakness of their financial environment, which have encouraged the inefficient allocation of savings and led in turn to the negative growth in real GDP.

3 The results on this section are not reported in the paper to keep the number of tables to a minimum, but they can be obtained from the author upon request.

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

149

Acknowledgments I wish to thank the editor, an anonymous referee of this journal, and E. Philip Howrey of the University of Michigan for helpful comments and constructive suggestions on an earlier draft of this paper. The responsibility for all errors is, however, mine.

References
Bencivenga, V. R., & Smith, B. D. (1991). Financial intermediation and endogenous growth. Review of Economic Studies, 58, 195 – 209. Bishop, R. V. (1979). The construction and use of causality tests. Agricultural Economics, 31, 1 – 6. Darrat, A. F. (1999). Are financial deepening and economic growth causally related? Another look at the evidence. International Economic Journal, 13 (3), 19 – 35. Demetrides, P. O., & Hussein, K. A. (1996, December). Does financial development cause economic growth? Time-series evidence from 16 countries. Journal of Development Economics, 387 – 411. Enders, W. (1995). Applied econometrics time series. New York: Wiley. Engel, R. F., & Granger, C. W. (1987). Cointegration and error-correction representation, estimation, and testing. Econometrica, 55, 251 – 276. Friedman, M., & Schwartz, A. J. (1963). A monetary history of the United States. Princeton. NJ: Princeton University Press. Gertler, M., & Rose, A. (1991, December). Finance, growth, and public policy. Working Paper WPS 814. Washington, DC: The World Bank. Gilber, C. L. (1986, August). Professor Hendry’s econometric methodology. Oxford Bulletin of Economics and Statistics, 283 – 303. Goldsmith, R. (1969). Financial structure and development. New York: Yale University Press. Granger, C. W. (1986). Developments in the study of cointegrated economic variables. Oxford Bulletin of Economics and Statistics, 213 – 228. Granger, C. W., & Newbold, P. (1974, July). Spurious regressions in econometrics. Journal of Econometrics, 111 – 120. Greenwood, J., & Jovanovic, B. (1990). Financial development, growth, and the distribution of income. Journal of Political Economy, 98, 1076 – 1107. Greenwood, J., & Smith, B. D. (1997, January). Financial markets in development and the development of financial markets. Journal of Economic Dynamics and Control, 145 – 181. Gregorio, J. D., & Guidotti, P. E. (1995). Financial development and economic growth. World Development, 23 (3), 433 – 448. Ireland, P. N. (1994, March). Money and growth: an alternative approach. American Economic Review, 47 – 65. Johansen, D. (1988). Statistical analysis of cointegration vectors. Journal Economic Dynamics and Control, 12, 231 – 254. King, R. G., & Levine, R. (1993). Finance and growth: Shumpeter might be right. Quarterly Journal of Economics, 108, 717 – 737. Levine, R. (1997). Financial development and economic growth: views and agenda. Journal of Economic Literature, 35 (2), 688 – 726. Lucas, R. E. (1988, July). On the mechanics of economic development. Journal of Monetary Economics, 3 – 42. McKinnon, R. (1973). Money and capital in economic development. Washington, DC: Brookings Institution. Pantula, S. G., Gonzalo-Farias, G., & Fuller, W. A. (1994, October). A comparison of unit-test criteria. Journal of Business and Economic Statistics, 449 – 459. Robinson, J. (1952). The generalization of the general theory. The rate of interest and other essays. London: Macmillan.

150

Y. Khalifa Al-Yousif / Review of Financial Economics 11 (2002) 131–150

Shaw, E. (1973). Financial deepening in economic development. New York: Oxford University Press. Shumpeter, J. A. (1932). The theory of economic development. Cambridge, MA: Harvard University Press. Stock, J., & Watson, M. (1989). Interpreting the evidence on money-income causality. Journal of Econometrics, 40, 161 – 182. Thakor, A. V. (1996). The design of financial systems: an overview. Journal of Banking and Finance, 20, 917 – 948. Vogel, R., & Burser, S. (1976). Inflation, financial repression, and capital formation in Latin America. In McKinnon R. (Ed.), Money and finance in economic growth and development. New York: Marcel Dekker, Inc. Wang, E. C. (1999). A production function approach for analyzing the finance-growth nexus: the evidence from Taiwan. Journal of Asian Economics, 319 – 328. World Bank (1993). The Asian miracle: economic growth and public policy. New York: Oxford University Press.


相关文章:
Financial development and economic growth_ Recent e....pdf
Financial development and economic growth_ Recent evidence from China_金融/投资_经管营销_专业资料。Journal of Comparative Economics 40 (2012) 393412 ...
Financial Development and Economic Growth in Underd....pdf
Financial Development and Economic Growth in Underdeveloped Countries_Patrick(1966)_经管营销_专业资料。Financial Development and Economic Growth in Under...
law finance and economic growth in china.pdf
law finance and economic growth in china_经济学_高等教育_教育专区。...0.60 (1.14)a N/A Financial development (banking and market sectors) ...
Financial Development and Economic Growth in Southern Africa..unkown
1 1. Introduction In the last two decades, the link between financial intermediation (FI) and economic growth has generated a great deal of interest ...
Financial development and economic growth: A review.unkown
Financial development and economic growth: A review Chee-Keong Choong1* and Sok-Gee Chan2 1Centre for Economic Studies, Faculty of Business and Finance,...
Financial Development and Economic Growth: Views and Agenda.unkown
688726 Levine: Financial Development and Economic Growth Financial Development and Economic Growth: Views and Agenda ROSS LEVINE University of Virginia I ...
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH*.unkown
CROATIAN ECONOMIC SURVEY 1996 - 1999 207 FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH* Velimir onje** 1 INTRODUCTION Literature which examines the relationship...
Financial development and economic growth.unkown
Financial development and economic growth Author: Iulia Elena Mihai Coordinator: Laura Obreja Braoveanu Abstract The debate on the relationship between ...
FinancialDevelopment and Economic Growth.unkown
POLICY RESEARCH WORKING PAPER 1678 FinancialDevelopment and Economic Growth Views and Agenda Ross Levine A growing body of theoretical and empiricalwork would ...
Financial Development and Economic Growth: A Review of ....unkown
Financial Development and Economic Growth: A Review of Literature Tridip Ray Introduction Relationship between financial development and economic growth is a ...
Financial Development and Economic Growth. Edition No. 1.unkown
Brochure More information from http://www.researchandmarkets.com/reports/1906741/ Financial Development and Economic Growth. Edition No. 1 Description: Prior...
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH : ASSESSING THE ....unkown
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH : ASSESSING THE EVIDENCE* Philip Arestis and Panicos Demetriades In this paper we take a fresh look at the ...
Financial Development and Economic Growth.unkown
Financial Development and Economic Growth B ritisll A ssociation fo r the A l f WII/(CIIICII( ofSdcl/(:c /lOoks Series Ed itor: Da vid Reb ...
Excess Financial Development and Economic Growth1.unkown
Excess Financial Development and Economic Growth1 Lorenzo Ductora, Daryna Grechynab aDeparment of Economics, University of Alicante, Spain bDeparment of ...
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH.unkown
CHAPTER V FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH 5.1 Introduction Growth Rates of Important Economic Variables 5.2 Test of Causality 5.3 Definitions of ...
Does Financial Developmentand Inflation Spur EconomicGrowth ....unkown
dynamics between financial development, inflation and economic growth during the post-1997 financial crisis in Thailand using battery of time series techniques....
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH.unkown
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH Friday 4th April 2008 TO REGISTER please RSVP by Tuesday 4th March, with details of your name, institution and ...
FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH IN NIGERIA.unkown
financial development and economic growth.2 In this paper, we also evaluate these hypotheses using appropriate methodology in order to ascertain the one that...
Puzzles in financial development and economic growth.unkown
Puzzles in financial development and economic growth Binam Ghimire* Gianluigi Giorgioni January 2009 Liverpool Business School, Liverpool John Moores University,...
Financial development and economic growth: convergence or ....unkown
Michael K. Fung 1 Abstract This study tests for convergence in financial development and economic growth by incorporating the interaction between the real...
更多相关标签: